Cache GroupByHash raw values where appropriate #25294
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pettyjamesm
force-pushed
the
cache-hash-values
branch
3 times, most recently
from
7c12b77 to
90f9484
Compare
dain
approved these changes
Mar 13, 2025
Adds logical to dynamically decide whether to cache hash values into the hash table itself based on whether the GroupByHash instance is spillable, whether container types are present in the grouping keys, the number of variable width types, and the number of types overall. Caching hash values in GroupByHash is a classic compute / memory trade-off. Caching the value costs more memory (8 extra bytes per record) but avoids relatively more expensive hash recalculation when re-hashing the table or sorting the table contents by raw hash value for spilling. Caching the hash value can also make inserting new entries cheaper by avoiding the need for relatively more expensive valueIdentical checks when the hash value can prove that the values are not identical, although this effect is already somewhat mitigated by the FlatHash control vector.
pettyjamesm
force-pushed
the
cache-hash-values
branch
from
90f9484 to
aa2e58f
Compare
Fair point, I've rephrased the commit message and PR description- although minor clarification, the control vector only stores 7 bits of the hash (since the sign bit is used to detect set vs unset). Thanks for the review! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Adds logic to dynamically decide whether to cache hash values into the hash table itself based on whether the
GroupByHashinstance is spillable, container types are present in the grouping keys, there are 2 or more variable width types present, or there are 3 or more types in total.Caching hash values in
GroupByHashis a classic compute / memory trade-off. Caching the value costs more memory (8 extra bytes per record) but avoids relatively more expensive hash recalculation when re-hashing the table or sorting the table contents by raw hash value for spilling.Caching the hash value can also make inserting new entries cheaper by avoiding the need for relatively more expensive
valueIdenticalchecks when the hash value can prove that the values are not identical, although this effect is already somewhat mitigated by theFlatHash#controlvector.Additional context and related issues
Builds on top of:
Release notes
( ) This is not user-visible or is docs only, and no release notes are required.
( ) Release notes are required. Please propose a release note for me.
(x) Release notes are required, with the following suggested text: